hushem_1x_beit_base_sgd_0001_fold5

This model is a fine-tuned version of microsoft/beit-base-patch16-224 on the imagefolder dataset. It achieves the following results on the evaluation set:

• Loss: 1.5504
• Accuracy: 0.2683

Model description

More information needed

Intended uses & limitations

More information needed

Training and evaluation data

More information needed

Training procedure

Training hyperparameters

The following hyperparameters were used during training:

• learning_rate: 0.0001
• train_batch_size: 32
• eval_batch_size: 32
• seed: 42
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• lr_scheduler_warmup_ratio: 0.1
• num_epochs: 50

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy
No log	1.0	6	1.6302	0.2439
1.5748	2.0	12	1.6252	0.2439
1.5748	3.0	18	1.6204	0.2439
1.5763	4.0	24	1.6160	0.2439
1.56	5.0	30	1.6118	0.2439
1.56	6.0	36	1.6079	0.2439
1.5722	7.0	42	1.6043	0.2439
1.5722	8.0	48	1.6006	0.2439
1.5053	9.0	54	1.5970	0.2439
1.5617	10.0	60	1.5937	0.2439
1.5617	11.0	66	1.5908	0.2439
1.5101	12.0	72	1.5876	0.2439
1.5101	13.0	78	1.5848	0.2683
1.5266	14.0	84	1.5820	0.2683
1.4925	15.0	90	1.5796	0.2683
1.4925	16.0	96	1.5771	0.2683
1.5202	17.0	102	1.5750	0.2683
1.5202	18.0	108	1.5729	0.2683
1.5168	19.0	114	1.5711	0.2683
1.5066	20.0	120	1.5691	0.2683
1.5066	21.0	126	1.5674	0.2683
1.508	22.0	132	1.5658	0.2683
1.508	23.0	138	1.5642	0.2683
1.4868	24.0	144	1.5626	0.2683
1.5018	25.0	150	1.5612	0.2683
1.5018	26.0	156	1.5598	0.2683
1.5061	27.0	162	1.5585	0.2683
1.5061	28.0	168	1.5574	0.2683
1.4922	29.0	174	1.5565	0.2683
1.5131	30.0	180	1.5557	0.2683
1.5131	31.0	186	1.5547	0.2683
1.5054	32.0	192	1.5540	0.2683
1.5054	33.0	198	1.5533	0.2683
1.4665	34.0	204	1.5527	0.2683
1.5093	35.0	210	1.5521	0.2683
1.5093	36.0	216	1.5516	0.2683
1.5042	37.0	222	1.5513	0.2683
1.5042	38.0	228	1.5509	0.2683
1.4952	39.0	234	1.5507	0.2683
1.4728	40.0	240	1.5505	0.2683
1.4728	41.0	246	1.5504	0.2683
1.4831	42.0	252	1.5504	0.2683
1.4831	43.0	258	1.5504	0.2683
1.4991	44.0	264	1.5504	0.2683
1.4929	45.0	270	1.5504	0.2683
1.4929	46.0	276	1.5504	0.2683
1.5005	47.0	282	1.5504	0.2683
1.5005	48.0	288	1.5504	0.2683
1.4392	49.0	294	1.5504	0.2683
1.4753	50.0	300	1.5504	0.2683

Framework versions

• Transformers 4.35.2
• Pytorch 2.1.0+cu118
• Datasets 2.15.0
• Tokenizers 0.15.0